In light of the negative consequences associated with the development of distress in initially satisfied marriages, the proposed research examines the structure of spouses' representations of the marriage for its role in the resiliency of marital satisfaction over time. Perceptions of a marriage are presumed to vary in breadth, from specific observations of partner behaviors to global evaluations of the relationship as a whole. The central hypothesis of this research is that the way specific perceptions of the partner are organized will moderate the association between those perceptions and global evaluations of the marriage. The more complex (i.e., differentiated and evaluatively integrated) the organization of specific perceptions, the less any one negative perception should impact global evaluations. Thus, spouses whose representations of their partners are more complex should experience more stable global satisfaction over time, exchange more positive behaviors during marital interaction, and adapt more successfully to stressful events, compared to spouses whose representations are less complex. To address several hypotheses derived from this model, 150 newlywed couples will be assessed every six months for the first four years of their marriage. Assessments will include: (a) marital interactions and structured interviews recorded in the laboratory shortly after marriage and two years later; (b) a week of nightly evaluations of global and specific perceptions of the relationship collected shortly after marriage and two years later; (c) self-report questionnaires administered through the mail at 6-month intervals. Hypotheses will be tested using growth curve analysis. To the extent that the structure of spouses' cognitions may prove easier to change than the content of those cognitions, the long-term goal of this research is to identify a potentially important but currently overlooked target for future interventions designed to prevent the deterioration of initially satisfied marriages.